ArcCollector 2: Project Activity

Introduction

          With experience in collecting data in ArcCollector for the microclimate, this activity had individuals plan and implement their own project through ArcCollector. The objective of this project was to develop a model for people to document litter in their surroundings. Trash has only become a bigger problem as time goes on. Besides not looking nice, improperly disposed-of trash can leech harmful chemicals into the environment, harm or kill wildlife, attract disease, and increase general hazards to people. By documenting litter, high litter patterns can emerge and professionals can utilize such information to implement solutions, preventative measures, and clean-ups (Figure 1).

Figure 1: A potential problem area on upper campus where the trees snag fly-away plastic bags.

           Before data collection is even possible, it's crucial to carefully plan and set up the project. Without a properly designed project, a collector will run into problems in the field and in the analysis, like false or inaccurate data points and unavailable documentation elements. In addition, a properly designed project will give accurate, relevant, and informative results.

Study Area

           At, first the study area was established to include UW - Eau Claire campus, Downtown, Owen Park, and a small neighborhood as shown by the red boundary lines in Figure 2. UW - Eau Claire includes upper and lower campus, Putnam Drive, and the property north of the Chippewa River. Downtown included South Farwell Street, Barstow Street, Graham Avenue, Wilson Park, and its perpendicular streets. The neighborhood connected Downtown to lower campus, bordered by the Chippewa River and State Street These study areas were meant to represent different types of spaces - campus, downtown, residential, and park spaces. However, the time to cover the total area was underestimated and made difficult given the weather, minimal daylight hours, and weak phone battery. Therefore, data was only collected in Downtown and the Campus property study areas. The objective of the activity still remains intact - collect litter data for documentation and analyzation. 

Figure 2: An image from ArcCollector of the different study areas and the where data collection took place.

Methods

          To begin the project, a file geodatabase was created in a personal folder. From ArcCatalog, the geodatabase's domains are established within the properties window (Figure 3). The domains chosen were - 

Amount - the number of articles of trash per approximate square meter

Bin - to designate whether a data point is a trash bin, recycling bin, compost, or no bin (thus a litter     data point)

Description - what type of trash it it

Area - downtown, residential, campus property, or park where the trash was found

Notes - a domain to mention relevant information of a point

          The domain must have a Field and Domain Type. This sets the boundaries for what type of information is recorded and can limit the possible errors made. Amount set as a Short Integer Field Type and a Range Domain Type. The rest were Text Field Types. The Text Fields were coded values in order to set up categories within the Bin, Description, and Area Domains. The Bin had values of Trash, Recycling, Compost, and No Bin. This way, trash, recycling, and compost bins could be documented too. If a litter point were to be uploaded, No Bin would be selected for its attribute. The Description Domain was set up as a coded value in an attempt to formally categorize the types of litter. The Area Domain also had coded values - residential, campus property, park, and downtown - in order to record the type of region the point was taken in (Figure 4).

Figure 3: The domain established in the geodatabase.

          

Figure 4: The coded values for the Bin, Area, and Description Domains that formed the categories in ArcCollector. Note, Graham Avenue was accidentally called Grand Avenue.

          Next, a feature class was created in the geodatabase called Litter and Bins. In the feature class properties window, the fields that will be apart of the feature class attributes are determined from the domains of the geodatabase (Figure 5).

Figure 5: The feature class properties window where the domains that will be used are established.

            Once the feature class was set up, it was published to ArcGIS online. In ArcGIS online bring in the  published layer onto a basemap. Nothing shows up because there are no data points in the layer. The map is shared and ready to be viewed in ArcCollector, and the fields that were defined for the feature class will be show up when plotting a point (Figure 6).

Figure 6: The fields that will be filled out according to the rules set up in the Domains when plotting a litter or bin point.

Results

       

          The first map shows the location of trash and recycling bins within the study areas (Figure 7). Downtown has a lot of trash bins but no recycling bins. On the other hand, the college campus displays plenty of trash and recycling bins. The recycling bins were usually attached to the trash bins, so the recycling bin points often overlaid the trash bin points on the map.

Figure 7: Trash and recycling bin locations.

          The second map shows the bin points along with the points that designate litter (Figure 8). There appears to be more litter where there are less trash bins in Downtown, and over all, at a higher density than on the college campus.

Figure 8: Trash and recycling bins and litter data points.

          The third map displays the Amount domain, or, the number of articles of trash per square meter (Figure 9). Downtown has the highest density of trash, sporting points in the 16-50 range. This can be attributed to the patches of cigarette butts deposits that weren't really found on campus.

Figure 9: Litter displayed as a graduated symbols map showing the amount of trash articles per square meter.

          The fourth map in the Litter Report displays information on the different categories of litter for each point (Figure 10). No pattern really jumps out except that Paper and Food Wrapping seem to be the two top categories on the college campus.

Figure 10: Litter displayed as its category.

          The fifth map for the Litter Report shows the type of area that the litter was found (Figure 11). Clearly, Downtown and Campus-defined points are located in their previously defined zones. However, the some litter points in the Downtown study area are defined as being located in a park as that study did contain Wilson Park. Upper campus could potentially be labeled as a residential area because that is where the residence halls are located. They ended up constituting campus property since it is and it's not a traditional neighborhood.

Figure 11: Litter displayed as the location that it was found in.

Conclusion

           Even with the knowledge of how important a proper project design is, mistakes were still made in this project. For some reason, the Notes domain was set up for coded values that weren't there, instead of a plain text field to freely record notes. Thus, no notes were recorded. It was discovered that cigarette butts needed their own category base on their prevalence, and they were lumped into the Other category. The Assortment category was used more than desired because there were many instances where a data point couldn't be properly described from the variety of trash, making the Description domain useless sometimes. The Paper products category was used to gather data on potentially recyclable products, but it ended up including many non recyclable products, like tissues and napkins. There should have been a category for disposable drink containers (Figure 12). The Food Covering category was meant to cover everything from candy wrappers to take-out containers, but the extent of what fell under this category warrants more subdivisions. While the original study area was too large for this project, its design isn't necessarily meant to be contained to these study areas. It's main idea of documenting litter in relation to area and bins for further analysis and plans of action can be implemented anywhere. However, in an ideal situation, consumption and waste needs to significantly decrease to ultimately combat this environmental crisis.

Figure 12: A disposable cup that seemed to lack its own category.